Model Configuration

Nexu supports two model paths:

Nexu Official: platform-managed access with no provider key stored by the user
Bring your own key: connect a provider such as Anthropic, OpenAI, Google AI, or a custom OpenAI-compatible endpoint

Supported provider patterns

Provider	Typical base URL	Key format
Anthropic	`https://api.anthropic.com`	`sk-ant-...`
OpenAI	`https://api.openai.com/v1`	`sk-...`
Google AI	`https://generativelanguage.googleapis.com/v1beta`	`AIza...`
Custom	your OpenAI-compatible endpoint	provider-specific

In the Nexu UI

Open the Models page.
Choose a provider.
Paste the API key.
Optionally set a custom base URL.
Verify the connection.
Enable the models you want available in the workspace.

Operational guidance

Prefer least-privilege provider keys.
Do not share provider keys in screenshots, support tickets, or git history.
Verify before enabling a provider so Nexu can discover available models.
Use a custom provider when you need a proxy, self-hosted gateway, or compatible inference service.

Default examples

Nexu currently highlights these common model families:

Anthropic: claude-opus-4, claude-sonnet-4, claude-haiku
OpenAI: gpt-4o, gpt-4o-mini, o1, o3-mini
Google AI: gemini-2.5-flash, gemini-2.5-pro

Best practice

Use Nexu Official for the fastest onboarding, then add BYOK providers only when you need model control, billing separation, or custom routing.